guided questionnaire for the development of a sswl type … · 2019. 5. 2. · 1 guided...

Guided questionnaire for the development of aSSWL type database of Adjectival Properties in

the World’s Languages

Hilda KoopmanDepartment of Linguistics, UCLA

[email protected]

version October 31 2014This questionnaire is in development

comments welcome, collaboration welcomed,(there are typos, and open ended sections ..)

Contents1 Guided questionnaire for the development of a SSWL type

database for Adjectival Properties 41.1 Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41.2 Outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51.3 What the questionnaire does not do . . . . . . . . . . . . . . . . 51.4 (Some) Background . . . . . . . . . . . . . . . . . . . . . . . . . 51.5 A brief note on the organization of the questionnaire . . . . . . . 6

1.5.1 How to use the questionnaire . . . . . . . . . . . . . . . . 7

2 Semantic interpretations 82.1 Intersective Adjectives . . . . . . . . . . . . . . . . . . . . . . . . 82.2 Nonintersective adjectives . . . . . . . . . . . . . . . . . . . . . . 9

2.2.1 Follow-up questions and tables: non-intersective adjec-tives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

2.2.2 a good N vs be good . . . . . . . . . . . . . . . . . . . . 132.2.3 An aside: fake.. . . . . . . . . . . . . . . . . . . . . . . . . 13

3 "Descriptive" adjectives – Basics 153.1 Adjectives and adjective-like categories: Semantic classes and

forms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

[email protected]

3.1.1 sideline: bound morphemes . . . . . . . . . . . . . . . . . 163.2 Predicative As . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

3.2.1 Basic Vocabulary: Adjectives from the Swadish list . . . . 183.3 Reduced relatives: verbal adjectives/participles . . . . . . . . . . 193.4 "True" adjectives . . . . . . . . . . . . . . . . . . . . . . . . . . . 193.5 Questions about Form, order, and Modification (1) . . . . . . . . 19

3.5.1 Forms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193.6 Indefinite noun phrases . . . . . . . . . . . . . . . . . . . . . . . 19

3.6.1 Order . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193.6.2 Modification . . . . . . . . . . . . . . . . . . . . . . . . . 203.6.3 Noun class/gender and sg nouns: indefinites . . . . . . . . 213.6.4 Singular/plural marking . . . . . . . . . . . . . . . . . . . 21

3.6.4.1 Summary Table for plural indefinites . . . . . . 223.6.5 Summary table: forms . . . . . . . . . . . . . . . . . . . . 22

3.7 Definite noun phrases . . . . . . . . . . . . . . . . . . . . . . . . 233.7.1 Multiple articles/linkers . . . . . . . . . . . . . . . . . . . 243.7.2 Full relative clauses (tensed finite relatives) . . . . . . . . 25

4 Combining adjectival fields: orders with respect to the N, rel-ative orders of adjectives, and tables 264.1 Basic orders etc: Indefinites . . . . . . . . . . . . . . . . . . . . . 26

4.1.1 Table: General Order w.r.t to N in Indefinite noun phrases 274.1.2 Table: General Order w.r.t. N: definites . . . . . . . . . . 27

4.2 Relative Ordering of adjectival regions: a full paradigm withfollow-up tables) . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

4.3 Subjective comment with all other regions . . . . . . . . . . . . . 284.3.1 Subjective comment and size . . . . . . . . . . . . . . . . 284.3.2 Subjective comment and age . . . . . . . . . . . . . . . . 294.3.3 Subjective comment and shape . . . . . . . . . . . . . . . 294.3.4 Subjective comment and color . . . . . . . . . . . . . . . . 304.3.5 Subjective comment and gender . . . . . . . . . . . . . . . 314.3.6 Subjective comment and nationality . . . . . . . . . . . . 314.3.7 Subjective comment and material . . . . . . . . . . . . . . 32

4.4 Combining Size with the remaining regions . . . . . . . . . . . . 334.4.1 size and age . . . . . . . . . . . . . . . . . . . . . . . . . 334.4.2 size and shape . . . . . . . . . . . . . . . . . . . . . . . . 334.4.3 size and color . . . . . . . . . . . . . . . . . . . . . . . . 344.4.4 size and gender . . . . . . . . . . . . . . . . . . . . . . . 354.4.5 size and nationality . . . . . . . . . . . . . . . . . . . . . 354.4.6 size and material . . . . . . . . . . . . . . . . . . . . . . . 36

4.5 Combining age with remaining regions . . . . . . . . . . . . . . . 374.5.1 age and color . . . . . . . . . . . . . . . . . . . . . . . . . 374.5.2 age and gender . . . . . . . . . . . . . . . . . . . . . . . . 374.5.3 age and nationality . . . . . . . . . . . . . . . . . . . . . 384.5.4 age and material . . . . . . . . . . . . . . . . . . . . . . . 39

4.6 Combining color with remaining regions . . . . . . . . . . . . . . 39

2

4.6.1 color and gender . . . . . . . . . . . . . . . . . . . . . . . 394.6.2 color and nationality . . . . . . . . . . . . . . . . . . . . 404.6.3 color and material . . . . . . . . . . . . . . . . . . . . . . 41

4.7 Combining nationality with remaining regions . . . . . . . . . . . 424.7.1 nationality and material . . . . . . . . . . . . . . . . . . . 42

4.8 Definites noun phrases . . . . . . . . . . . . . . . . . . . . . . . . 43

5 Order: Num A N 445.1 Order: Num A N indefinite . . . . . . . . . . . . . . . . . . . . . 445.2 three adjectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445.3 Order: Num A N definite . . . . . . . . . . . . . . . . . . . . . . 45

6 Orders: Dem A N 45

7 Universal 20: Greenberg (1965) and Cinque (2005): the relativeorders of Dem Num A N 467.1 Table Dem initial . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

7.1.1 follow up questions Dem initial . . . . . . . . . . . . . . . 467.2 Table N initial . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

7.2.1 follow up questions N initial . . . . . . . . . . . . . . . . . 477.3 Table A initial . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

7.3.1 Follow up questions A initial . . . . . . . . . . . . . . . . 477.4 Table: Num initial . . . . . . . . . . . . . . . . . . . . . . . . . . 47

7.4.1 Follow up questions Num initial . . . . . . . . . . . . . . . 487.4.2 Other . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

8 To do: here is an incomplete list of still missing from question-naire, and appendix: feel free to add using the formats given 48

A MOVE: Further questions about reduced relative clauses 49A.1 Order: N red Rel/Red rel N . . . . . . . . . . . . . . . . . . . . . 49A.2 Order Num RedRel N . . . . . . . . . . . . . . . . . . . . . . . . 49A.3 Order Num RedRel N Indef . . . . . . . . . . . . . . . . . . . . . 50A.4 Order: Ord RedRel N definite . . . . . . . . . . . . . . . . . . . 50A.5 Order Dem RedRel N . . . . . . . . . . . . . . . . . . . . . . . . 51

B Compounds [SKIP? ELABORATE] 51

C Further questions about modifiers: comparatives and superla-tives 52

D ideophones, from Russel Schuh 52

E References Cited 52

F Appendix: Snider et al (2004) for African languages 55

3

1 Guided questionnaire for the development of aSSWL type database for Adjectival Properties

1.1 GoalsOn the most general level, this questionnaire aims

• to (start to) generate finely-grained comparative data about the distri-bution of adjectives and adjective-like classes in the world’s languages,building, as much as possible, on current understanding emerging fromthe formal and typological literature of the (morpho) syntax, semantics ofadjectives and adjective-like classes.

• To generate tables that summarize the properties for individual languages,what is attested and what is not, in a format which allows easy enteringof the data into the open source sswl database SSWL on Terraling, thenext generation of the database http://test.terraling.com, thus graduallydeveloping a finely grained comparative database which can serve as abasic tool in support of scientific research.

To reach these goals, in-depth studies and descriptions of individual languagesare necessary, and these must involve native speaker linguists (to the extentpossible).The questionnaire should serve as a guide for generating data, and thedescription (and analysis) for individual languages. It should give clear, stepby step instructions on how to proceed, so as to be able to code what is foundin a particular language, and what is not. The basic idea is to break the tasksinto small manageable pieces: given the complexity involved, there will be verymany of these tasks, each of which should be relatively self contained.The questionnaire will without any doubt go through revisions, and furtherexpansions.

• The questionnaire is a mix between the traditional linguistic questionnairemodel, with the current property definitions of SSWL (see website foradjectival properties SSWL (these give a specific definition, define thetask, give specific instructions for elicitation contexts, exemplify how theproperty is set for several languages (using yes and no values), and askcontributors to decide this task for his or her language). The questionnaireis set up in such a way that the many small properties that are necessaryfor coding cross linguistic variation can be easily summarized into Tables,which follow each subsection. Once the basic data and examples have beengathered, native speaker linguists are asked to code the relevant propertiesas yes/ no/ not sure values. These tables and the examples that illustratethem will constitute the basic data for the sswl database.

• For ease of entry, a separate excel file with the tables is under construction.

4

http://test.terraling.com/groups/7http://test.terraling.comhttp://test.terraling.com/groups/7

1.2 Outcomes• Presentations and Publications: In depth descriptions and/or analyses of

individual languages, to be published as a companion to the database onthe SSWL site.

• Tables for the database that summarize the properties for individual lan-guages, and many as are needed, with examples illustrating the properties,to be checked and entered into the sswl database (tagged by name of con-tributor).

• Presentations and publications on comparative syntax for closely relatedlanguages (Cameroon,..)., or for broad typological purposes.

1.3 What the questionnaire does not do• The questionnaire does not guide a native speaker linguist in how to put

the description into a narrative: this should be guided by the general-izations that emerge from the data. The native speaker linguist will ofcourse be able to restrict the description to some subset of data in thequestionnaire, while still contributing to the broader set of data.

• The questionnaire does not show how to pursue analyses: to some extent,tests are built into the questions, in such a way that some hypotheses aboutthe fine structure of adjectival regains can be answered. Any analysisstarts with questions, and finding ways to answer them, i.e. applyingor setting up tests that can be used to probe the structures. Here the(vast) literature should be studied and consulted of course. Any particularquestion in this vast domain can be pursed in depth: this is up to theindividual researcher(s).

1.4 (Some) BackgroundThe literature on adjectives by now is vast, both in the typological tradition,Greenberg (1965), Dixon (2004), Rijkhoff (2002), etc, which draws on datafrom many languages, and in the formal syntax and semantics tradition whichis based on a large number of in depth studies on particular languages over theyears. This questionnaire draws in particular on Cinque (2010), whose bookThe syntax of adjectives: a comparative study builds on decades of (his own)research on adjectives in Romance and Germanic languages, incorporates theformal syntactic and semantic literature, and the findings of the vast linguisticliterature on adjectives in the typological tradition (the reference section spans36 pages, and is by itself a valuable source). The resulting picture suggestsremarkable crosslinguistic stability from a bewildering complex and opaquepicture of adjectives and adjectival-like classes in each individual language,and makes clear predictions about the expected behavior and interpretationdepending on relative orders, and environments in which the adjectives occur.

5

One important finding by Cinque is that different semantic interpretationssystematically seem to correlate with specific adjectival regions as revealed bythe syntactic structures.In this questionnaire, this is exploited to some degree, through the importantdifference between intersective adjectives and nonintersective adjectives, andwhat adjectival region the latter unambiguously must enter in (the regionCinque calls direct modification), after Larson 2005 and others).

A second important finding is that at least in some languages, someadjectives can enter into more than one structural region, creating structuralambiguity, and much opacity. The diagnostics Cinque develops help clear upthis picture considerably, and make clear testable predictions about correlations(apparent flexible word order (with preferred orders), or fixed orders adjectives),interpretative properties (predicative/ non predicative), predictions about linearorder variations etc.

As we know from the typological literature (Dixon 2004, Rijkhoff 2002,among others), there is variation in the size of the class of adjectives (formvery many to few, or none), and there is variation in the formal realizationof the category "adjective" : whether they look like reduced relative clauses,verb-like, noun-like, true adjectives or compounds, how many classes are foundetc . Taken together with Cinque’s finding that a class of adjectives in Englishetc can appear , with no distinction in form, in two different regions, one forreduced relatives (indirect modification), the other closer to the noun (directmodification), the questionnaire will target adjectives and adjectival-like classestogether. Questions about form are built into the questionnaire. This willallow determine how individual languages lexicalize these (possibly havingmultiple classes), and prepare for possible test cases (see Cinque p. ++) totest if these regions occur crosslinguistically, or not, and if so if the hierarchicalarrangement is variable or invariant. This will allow moving towards a fine mapof the internal structure of adjectival regions through comparative data.

++DELETE THE FOLLOWING

1. What adjectives or adjectival classes do we find in a particular language?

2. What are their formal morpho-syntactic and interpretative properties?

3. What is the hierarchical arrangement of classes of adjectives; how manyadjectival ’fields’ are there; how are they structured internally; and howare these ordered w.r.t. the N and other elements in the noun phraseproperties (definite, indefinite), demonstratives, numerals, ordinals, dif-ferent sizes of relative clauses, ?

1.5 A brief note on the organization of the questionnaire• The questionnaire starts out in a somewhat unusual way, with a section

on the difference between predicative (roughly intersective), and nonpred-icative (nonintersective) adjectives modifying nouns. As a prerequisite for

6

probing certain structures, particular adjectives (non-predicative/non in-tersective ) need to be found for each language (if the language has themat all), to probe the structures. This section should be thought of asproviding basic lab tools that can provide important information aboutclasses of adjectives, the relation between their form, interpretation, andsyntactic realization.

– Start the questionnaire, by carefully read through the section on 2.2.Do this at least once. This should help you start looking for non-intersective adjectives. Do not worry if you feel this is too daunting:just keep it in mind, and return to this section at some later point.The expected section on the basic classes of adjectives (size, coloretc), forms, and distributions is found in section 3, and should bedone by everyone.Work your way through this section, enter your examples and com-ments into a separate file.

– Each section, throughout the questionnaire, has follow-up questions,and Tables which summarize the findings in that section, asking foryes, no, not sure values. (Simply check the rows for yes values, andnon sure values, in the questionnaire).Scan your pages with tables, and save them. (An excel file with allthe tables is under development).

1.5.1 How to use the questionnaire

This questionnaire can be used at many levels to explore adjectives andadjectival classes in a particular language, and generate the data that underlyany description or analysis. It is directed mainly to native speaker linguists.

It aims to code not only what we find in your language, but also whatwe don’t find, but do find in other languages, or might expect to find givencertain theories). If we do this for each language, and refining the tablesof variation ultimately to their smallest components, we will construct anever expanding database which can be queried to allow comparisons, testcorrelations, implications, and predicted gaps, and clustering of properties(which properties never seem to go together?).

Section 3 covers the basic classes (size, shape, color etc) (forms, classes ofadjectives, and adjectival like categories, agreement, order, etc), and is followedby Tables which should be filled out.Sections about orders should be filled out. Each section is followed by tableswhere you are asked to summarize the findings, by checking boxes, which asksfor yes/no/not sure values. Indicate at least yes and not sure values (thesecan be combined). These tables will serve to code for database purposeswhat is found and what is not found in your language. For database anyhuman language counts. This should allow to construct a database that allowscomparing closely related languages or even dialects. The plan is to publish

7

descriptions that come out of the questionnaire electronically on the SSWLsite. Minimally, all the ordering properties, and the tables relating to theseshould be filled out..

2 Semantic interpretationsThe semantic interpretation of adjectives falls into two major classes (inter-sective (i.e. predicative) adjectives and non-intersective adjectives (i.e. nonpredicative), with different subclasses). This section introduces the differencebetween intersective adjective and nonintersective adjectives. It is meant tohelp you find non-intersective (or non predicative) adjectives in your language.At this point, questions about the category label adjective are not central.

Why are nonintersective adjectives important?Non-intersective (i.e. non-predicative adjectives) have a more restricted syntac-tic distribution than (one class) of intersective adjectives. Read in particular(Cinque, 2010), and references cited there. This means that they be used totease apart certain apparently very similar structures, or allow determining ifsome orders (violations of Universal 20 (Greenberg’ s 1966) Cinque (2005) areviolations of some universal principle, or not.

Study this section, and familiarize yourself with understanding how toidentify intersective and non-intersective readings. This should help youunderstand what to look for in a particular language, and help you identifynon-intersective adjectives in an individual language: (you may find none, justa few, or many. )with time, working through the questionnaire and literature, you may think ofmore.

2.1 Intersective AdjectivesSome adjectives, as in (1) give rise to intersective interpretations.

(1) a big dog

(1) is true for x, where x is big and x is a dog, i.e it is the intersection of set ofindividuals that are big and the set of individuals that are dogs.

This reading can always be paraphrased by a relative clause structure, with theadjective as main predicate a dog which is big. In some languages, it may bethe case that the form of the main predicate cannot expressed in the same wayas the adnominal adjective. (look for Yoruba examples). This does not meanthe adnominal adjective is not intersective. It just means the form can onlycombine with a N.

(2) a. a big dogb. a dog who is big

8

2.2 Nonintersective adjectivesNot all adjectives (or adjectival-like classes) (in English, Italian, Chinese,..) give rise to intersective interpretations. There are different classes ofnonintersective adjectives, but they can all be easily recognized:

(3) Non intersective adjectives cannot be paraphrased as a relative clausewith the adjective as main predicate, keeping the meaning constant.

The distinction between predicative and non predicative adjectives is importantand runs through the questionnaire.

The following examples should help you identify what to look for to findnon-intersective adjectives in your language.

1. (former, past, old, mere, future ..)

(4) John is a former senator

does not mean John is former and John is a senator but rather: John wasformerly a senator/ John was a senator in the past.The paraphrase test fails: former senator cannot be paraphrased as arelative clause: *A senator who was former

2. oldThe readings for old in English are ambiguous:

(5) a. Mary is an old friend (intersective: a friend who is old)b. Mary is an old friend (non-intersective: a longtime friend)

(5) b cannot be paraphrased as a relative clause: *a friend who is longtime.In some languages, old may also be able to mean former (no longer afriend).

3. Some adjectives can give rise to either intersective or nonintersective mean-ings, depending on the structure.

(6) Mary is a beautiful dancera. [beautiful] [ dancer ]

= Mary is a dancer who is beautiful (intersective)b. [ [ beautiful danc] er ]

= Mary is someone who dances beautifully (adverbial, non-intersective)

It is not known at present how widespread such structural ambiguitiesare, in particular if they depend on linear order asymmetries or not, i.e.are the non-intersective (adverbial) reading also available in languages in

9

which the noun precedes the adjective: N A (lit dancer beautiful). Is theambiguity systematically available in languages with the A N order ornot?An aside: Task: COME BACK WHEN YOU ARE DONE WITH THECORE PARTS OF THE QUESIONNAIRETo help generate data on these questions, translate sentences like (6) inyour language, (give glossed examples), and inspect the possible readings.Discuss with other native speakers if you are not sure.Here are some further examples to check:

(7) he is a heavy smokerhe is a big smokershe is a well wisher

(8) a. Does your language show the same ambiguity as in (8)? Yes/No/not sure(potential follow up, does it depend on the type of adjective?answering this question will require of work)

b. What is the relative order in your language in these cases? AN, N A or both?

c. Possible follow up if your answer to a is Yes: In English theambiguity disappears when you add young, and beautiful herecan only have an intersective (i.e. predicative) reading : Maryis a beautiful young dancer? (see Saba (2008))

4. Idiomatic adjectives. cf. red herring does not mean a herring which isred, but rather something like "a false lead"// Idiomatic adjectives arenecessarily non-intersective, and seem to occur close to the noun, makingit sometimes hard to decide if they are compounds or simply in a regionclose to compounds (For some properties and tables of compounds, seeappendix B )Such adjectives can also enter into other adjectival regions, as well as(depending on the language), correlating with different interpretive prop-erties.The following triplet from Italian (Cinque 2010 p. 90) shows the sameadjective giallo ’yellow’ in different orders, correlating with different in-terpretative properties.

(9) a. a roman giallolit: a novel yellow ("a thriller")(idiomatic: non intersective, post nominal, close to the N, "di-rect modification")

10

b. le gialle colline dell’Andalusiathe yellow hills of Andalousianon-intersective, "stage level" ("which are presently in a yellowstate), prenominal, "direct modification"

c. (Passami) la cravatta giallaLit: (Pass me) me the tie yellow. (pass me the tie which isyellow) intersective, restrictive, postnominal, "indirect modifi-cation"

This is analyzed by Cinque as the adjective occurring in different hierar-chical layers of the two adjectival fields, a complex field of "direct mod-ification" adjectives (Larson 2005), and a field of "indirect modification"adjectives. The field of indirect modification adjectives is hierarchicallyordered outside the layers for direct modification adjectives. see Cinquefor further discussion.) This adjective can potentially enter into differ-ent layers of adjectival fields in the noun phrase. These fields seem tobe ordered hierarchically; with idioms closest to N in an inner layer ofstructure, adjacent or partially overlapping with the compounding region,direct modification adjectives in a different layer of structure, and purelyintersective A in the outer layer, which appears to be dedicated to reducedrelatives.

Indirect > &Direct > &Direct Mod>& NModification &Modification & ("idioms") & N(9-c) & (9-b)& (9-a)reduced rel" & .. color A.. & idiom & N

2.2.1 Follow-up questions and tables: non-intersective adjectives.

1. Look for nonintersective adjectives/adjectival-like classes, (or just identifypotential candidates. )Make a list.Give examples

2. Look for idiomatic adjective noun (or noun adjectives) in your language.Make a list.Give examplesThese will provide you with tests that test certain hypotheses, and mayserve you later in the questionnaire: combining these with other adjectives,and looking at ordering and interpretations, can potential provide impor-tant evidence about finer adjectival properties in your language. Think ofthese as lab tools that are part of what you need to analyze adjectives andtheir representations.Non-predicative (non-intersective As) will be important (for someconstellations of data)throughout the questionnaire

11

(10) a. Give the order(s) w.r.t. the N for nonintersective adjectives:(as many yes as are found, leave the table blank if you havenot found any yet)

Table 1:Order & yes & no & don’t knowAnonintersect N & & &N Anonintersect & & &

b. If you find idiomatic adjectives, how are they ordered withrespect to the N?

Table 2:Order & yes & no & don’t knowAidiom N & & &N Aidiom & & &

c. If you find both classes, try if you can combine nonintersectiveand idiomatic adjectives; If you can, describe the relativeordering properties.

Further follow-up questions. (you may want to return to this questionafter 3)

(11) a. If As in your language look like (reduced) relative clauses(section 3.3)are there any nonintersective or idiomatic adjectives that looklike relative clauses?.

yes/ no /not sure?If yes, give examples.

b. If not, are there any candidates for nonintersective in ad-verbs/tense particles etc? (perhaps past tense/ temporal ad-verbs, future markers (future president) particles


c. Can adjectives be used adverbially, as in (John runs fast)

yes/ no /not sure?

12

If yes, give examples.

d. Does your language build adverbs from adjectives, (e.g. En-glish: quick, quickly clever: cleverly,.. etc.


(12) Advanced:For an exploration of a long list of further interpretive propertiesand references, study Cinque (2010) chapter 2, and try to con-struct similar examples in your language (in as far as they apply).This is advanced and difficult material. You may want to thinkabout how you would say in your language: visible, different, only,other.

2.2.2 a good N vs be good

COMMENTS/FEEDBACK NEEDED?

(13) Mary is a good doctor.

does not mean: Mary is good and Mary is a doctor(e.g. Mary can be good as a doctor but bad as a person). It rather means thatMary is good as a doctor, or that Mary is good at being a doctor. This readingis called subsective.

The paraphrase test however does not seem to fail (in English): Mary is agood doctor can be paraphrased as a relative clause Mary is a doctor who isgood.Translate:Mary is good (as a person)), Mary is a good person..This doctor is good (as a doctor)

In some languages a nominal element seems to be obligatory with this typeof adjective.

(14) a. Can you say in your language, keeping the meaning constant, andwhere good does not look nominal:Mary is good? This doctor is good? Yes/No/ Not sure.Give examples.

b. Do you have to add a N (Mary is a good person/doctor/professor,or some nominal element like "one": Mary is a good one?)

2.2.3 An aside: fake..

COMMENTS NEEDED?

13

(15) This is a fake knife

Does not mean this is fake and this is a knife, but rather something like thislooks like a knife but this not a knife. 1 Check:Here the paraphrase test causes some problems; fake in English can -at leastappear to look predicative: ((16) a)

(16) a. This knife is fakeb. This knife is a fake

How would you translate (15) in your language?

1This meaning is sometimes called privative.

14

3 "Descriptive" adjectives – BasicsThis section established basic formal properties of adjectives, and adjectival-like classes Identify potential adjectives or classes of adjectival-like categories,depending on meaning, and classify these according to their morphosyntacticproperties.You may find many adjectives in your language,you may find very few, or youmay find different adjectival classes (noun like, verb like, or relative clause like).If you find different adjectival-like classes, you want to pursue over time whatproperties the different classes have in common, and how they differ.This section will walk you through basic properties; It is followed by Tablesthat summarize the findings.

3.1 Adjectives and adjective-like categories: Semanticclasses and forms

Find candidates for adnominal adjectives in the semantic classes given in table1. The questionnaire will refer to these particular semantic classes as adjectivalregions. The regions are arranged according to what is suspected to be auniversal hierarchy, with the regions on the right combining with the nounbefore the regions to their left, etc. i.e material As combine iwith the N beforenationality A. Some of the regions on the right may show properties similar tocompounding. Treat these as a special class of adjectives for the purposes ofthe questionnaire.

Table 3:subjective comment& size&age&shape&color&gender&nationality&materialbeautiful & small & new/old & round &red & female & French & wooden &&..

• Construct examples, keep the contexts stable, and vary just the adjective.(This section deals with indefinite noun phrases only; for definite nounphrases, see section3.7)

• Make sure you combine the "adjective" with a noun: this insures the struc-ture is a noun phrase with an adnominal adjective rather than a sentence(with a silent copula/verb).

Here is an example from English:

(17) context: Did you build something this year?Answer:Yes, –

15

a. I built [ a nice house ]b. I built [ a small house ]c. I built [ a new house ]d. —- a round house; a French housee. etc.

• Give glossed examples, using the following format

(18) example sentenceglosstranslationcomments

• (Gradually) Classify adjectives according to their morphosyntactic prop-erties and semantic classes. (keep going back to your list, when you workthrough the questionnaire.)This will serve as the basis for follow up questions forms and tables (forindefinite, definite noun phrases, and relative orders of multiple adjectives.3.5, 3.6.5, and definite noun phrases 3.7

• Construct a basic table, with a list of adjectives and adjective like elements.

3.1.1 sideline: bound morphemes

• If certain adjectival concepts are realized as bound morphemes, (diminu-tive (i.e. small), or augmentative (i.e. big), treat them as a special classof adjectives, and answer basic questions about form:

(19) a. Do these bound morphemes precede or follow the N, or doesone part precede the N and one part follow?

b. Do these noun classes belong to a special noun class/gender?c. If there is agreement in your language, what kind of agree-

ment do these bound morphemes trigger?agreement with the N, agreement triggered by the augmen-tative (big) morpheme, or diminutive morpheme, or both.

3.2 Predicative AsFor each adjective in Table 2 in your language: check if it can be used as themain predicate in a sentence or not (20).

(20) (Question: Do you like the house?)Answer: Yes the house is nice;yes, it is niceyes, the house is big etc.

16

(21) (Question: What color is your bike?)Answer: My bike is red.etc.

Conclusions: nice and big etc can be used as the main predicate in a sentence.The form of the adjective is identical for adnominal and predicative adjectives.Compare the forms of adnominal adjectives, and predicative ones, and answer

the following questions: (these questions prepare for Table 6)

• is there a difference in form between the adnominal adjective and thepredicative adjective ?

• If yes, is there a shared part (a root)?

• Do adnominal adjectives (or some adjectives) look like nouns?

• Do predicative (or some) adjectives look like Ns?

• Do adnominal adjectives look like Verbs? In what way?

• Do predicative adjectives look like verbs? In what way?

• Do adnominal adjectives agree?

• Do predicative adjectives agree?− > If you find a difference in form, this can potentially be used to teaseapart cases of structural ambiguity. (The questionnaire follows up on thispoint at various places below).

17

3.2.1 Basic Vocabulary: Adjectives from the Swadish list

For purposes of historical reconstruction, it is useful to give the forms forthe following "adjectives" (use IPA fonts, whenever possible), drawn from theSwadish list or similar lists.Give as many forms per item that is attested in your language

Table 4:& adnominal & predicative& a good person & to be goodgood & &big & &small & &long (not ’wide’) (rope)& &red (color)& &black (color)& &green (color)& &yellow (color)& &white (color)& &full & &new& &round & &dry (substance) & &bitter& &

18

3.3 Reduced relatives: verbal adjectives/participles• Does your language have reduced relatives (e.g. verbal adjectives/participles)

of the following type?

(22) a. There is grilled fish on the menu(grilled= reduced relative, i.e. fish (which is) grilled))

b. fried chickenc. steamed riced. boiled cassavase. a painted wall

etc..

Compare the forms in (22) with a full relative clause:

(23) The chicken that I will fry tonight, ...

• Do any of these forms look like full relative clauses (i.e. do they share amorpheme like a relative clause marker)? Yes/ No/ Not sure

• What is the order with respect to the N? (N RedRel/ RedRel N), bothorders?

3.4 "True" adjectivesDoes your language also have ’true" adnominal adjectives: these could be quitesmall in number! and could exist next to nominal forms or verbal adjectives.NB these could be classified in descriptive grammars as compounds. (seeappendix for further questions about compounds)Even if this is the case, give the list. Give the list.

3.5 Questions about Form, order, and Modification (1)3.5.1 Forms

• Forms:Describe the forms of the (different classes of) adjectives (do they looklike adjectives, do they look like nouns, like verbs, or do they differ fromverbs and nouns, how?) (cf prepares for Table 3.6.5 and 6)

3.6 Indefinite noun phrases3.6.1 Order

How are adjectives (or the different classes of adjectives) ordered w.r.t. the N(indefinite noun phrases, no contrast).NB: As many yes as apply, i.e. code all attested orders for indefinite nounphrases. (there is no notion of dominant order, for the database we want to

19

need variation where it occurs)

Table 5: Order A w.r.t the N (Indefinites)

Order & yes & no & don’t knowA N & & &N A & & &

(24) a. If both orders are possible, does it depend on the type of adjectiveor not?

b. If both orders are possible with the same adjective and form, doesthe form of the adjective vary? If yes, describe how.

c. For each class of adjective-like elements: are the ordering possi-bilities the same or different: yes/ no. If no, explain what theorders are for each class of additives. True adjectives, verbal ad-jectives/reduced relatives, nominal adjectives.

3.6.2 Modification

Examine the list of adjectives/ adjectival classes found in your language

subjective comment& size&age&shape&color&beautiful & small & new/old & round &red &&

• For each class that is attested, can adjectives be modified by elements likevery (a very beautiful flower) ?

• If some As (or classes of As) can be modified, but not all, state whichones.Show examples

• Form: For each adjective-like class that is attested, can adjectives bemodified?

(25) a. A very beautiful cityb. a completely red wall

(26) A very beautiful/red/small/ bitter N

• For some further follow-up questions about comparatives and superlatives,see appendixC

20

3.6.3 Noun class/gender and sg nouns: indefinites

Vary the N for noun class (gender/animacy) (for each class of adjectives in(20)), for singular indefinite noun phrases):

(27) "a small person/amimal/knife/tree." "a black amimal/knife/tree."(28) Questions about noun class and singular indefinite noun phrases:

a. Does the form of the A vary with noun class (i.e. agree with thenoun)If yes, show and describe the forms (prefixes, suffixes, both, ..redu-plication,..)

b. Do all As change form? If not, does it depend on the linear orderA N, N A, or the classes of As in 3 (for example, it could be thatcolor adjectives don’t agree)?

Agreement:Does the form of the A vary with noun class in indefinite contexts? How?Do all A agree for noun class?How is agreement expressed?

3.6.4 Singular/plural marking

For each class of adjectives (where relevant), vary the N for singular/pluralmarking:

(29) a small child /small childrena big animal, big animalsa beautiful tree/ beautiful treesa long rope, long ropesa red ball/ red balls etc

(30) a. How are singular and plural expressed (indefinite contexts)(Is singular overly marked? if so how? Is plural marked, and how?(they may be expressed together)

b. Does singular/plural vary with noun class or gender?(Yes for noun class languages of the bantu type).(SKIP TOO DIFFICULT for other languages?))

c. Are singular and plural marked on both the noun and the A? (i.e. isthere visible agreement on at least some As in indefinite contexts?)(i) If so, do all classes of A agree?(ii) If there is more than one A, does each A agree?(iii) Plural marking: If plural is marked, is it marked once or

twice. If a linear order can be established, where does pluralappear?

21

3.6.4.1 Summary Table for plural indefinites

(31) Based on the properties so far, fill in the following table for pluralmarking, checking as many ’yes’ as applicable, illustrate each yes valuewith an example; for cases you are unsure about, check the Not surebox. No need to check the No box. Write No plural in indefinite NPsfor languages where the plural is unmarked in indefinitesAdd missing rows where necessary..(If singular marking is different, repeat the table for singular. ) Ifdefinite is different, repeat the table for definites. (or use a differentcolor for definites)

Plural marking (indefinite noun phrases)& & Yes & No & Not sure &Example1 & A pl N & & & &2 & A N pl & & & &3. & pl A N & & & &4.& pl A pl N & & & &5.& A pl N pl & & & &6. &pl A N pl & & & &7 & pl N A & & & &8 & N pl A & & & &9. & N A pl & & & &10.& pl N pl A & & &11.& N pl A pl & & &12. &pl N A pl & & &13. & N A A pl & & &14. & pl N A A & & &14. & N pl A A & & &15. & pl N pl A pl A & & &16. & pl N A A & & &17. & pl N pl A A & & &18. & pl N A pl A & & &19. & pl N A A pl & & &20. & pl A A N & & &21. & pl A A pl N & & &22. & pl A pl A pl N & & &&to complete & & &

3.6.5 Summary table: forms

Fill in the following table (yes/ no or not sure), based on your findings andanswers in the preceding sections. Check as many as applicable, (you may haveseveral classes of A in your language, this will come out in the table)) and copyor link the examples from your example file.

22

Check the not-sure column for any row you feel uncertain about, or ask forhelp (email!)Check as many as apply.

Table 6: Forms: Summary& && & yes& no & not sure& example1&(some) adnominal A and predicative A have the same form& & & &2 &(some)adnominal A and predicative A have a shared root & & &3 &(some) adnominal A look like a noun & & &4 &(some)adnominal A look like a verb & & &5 &(some)predicative A look like a noun & & &6 &(some)adnominal A looks like a Adj & & &7&(some) adnonimal A look like a relative clause& & &8 &(some) adnominal A shows agreement & & &9&(some) predicative A shows agreement & & &10 &predicative A can be modified & & &11 &(some) A can be modified & & &

*looks like a relative clause: e.g. has a morphosyntactic property that is sharedwith relatives (cf like a relative marker).Add comments (per row) (where relevant)Add rows if necessary. Additional tables for bare reduced relative clauses canbe found in the appendix.

3.7 Definite noun phrasesIn many languages, definite and indefinite noun phrases have different propertiesand internal structures (even in languages without definite articles).It is therefore very important to systematically control for definiteness. As afirst step, compare indefinite and definite noun phrases, and give examples:

(32) Context:a. I saw a small girl and a small boy . (First mention, indefinite)

Continuation:b. The small girl was laughing a lot. (Definite noun phrase)

(33) Context:a. I saw white birds and black birds Continuation:b. The black birds were sitting on a branch

(34) Context:a. I painted a red flower and a blue sky (indefinites)

Continuation:b. I like the red flowerc. The blue sky was nicer than the red flower

23

• Does the form of the A vary with the definiteness of the noun phrase?

• Does the order with respect to the noun vary, depending on definiteness?(cf. as in Shupamem, Nchare (2012) Yes/ No

• Is there a difference in phonology, prosody, or tones between definite/indefiniteNPS. ? Yes/No/ Don’t know.

• . Is there more than one "article/determiner"-like element, one on A andone on the nou n phrase in definite noun phrases?Is there an extra "article" /"augment/ linker" that appears on the A indefinite noun phrases? (the-big the-elephant)

• If the language marks definiteness where does the definiteness marker ap-pear? (ask for sswl property definitions excel sheet from Cristina Guardiano)You may also consider the following questions:

(35) a. If both orders are possible in definite contexts, does it dependon the type of adjective or not?

b. If both orders are possible with the same lexical item/semanticclass, does the form of the adjective vary? If yes, describehow.

c. For each class of adjective-like elements true adjectives, ver-bal adjectives, noun like adjectives..: are the ordering possi-bilities the same or different: yes/ no/ don’t know.If not, explain how they differ.

d. Can nonintersective adjectives like future or former, etc beused in definite contexts?

Fill in the following table for ordering of definite noun phrases w.r.t. theN. These could be identical in this simple case to indefinitenes, but theymight also vary.

Order & yes & no & don’t knowA N & & &N A & & &

3.7.1 Multiple articles/linkers

(36) a. Is there an indefinite article in your language?Yes/No /Not sure

b. If so, can there be more than one indefinite article in indefinitenoun phrase?a man a big/ a big a manYes/No /Not sure

24

(37) a. Is there is more than one article/definite determiner/linker(also know as determiner spreading, poly definiteness) in def-inite A N or NA combinations in your language?Yes, No, not sure examples

b. IMPORTANT If yes, can you have non-intersective adjectivesin definite contexts (go back to your list in section 2.2 , testfor each adjective, give examples. (Make sure these are finein indefinite contexts)

(38) If you have multiple determiners in your language (cf (38) a) isthis obligatory for all definite noun phrases, or does it depend onthe type of noun?Test what happens with unique objects, like sun and moon vswheather, stove..

(39) the hot sun vsthe hot weather/water/etc

Watch for whether the hot sun can be said in two ways (the equivalent tothe-hot the-sun vs. the hot sun with different meanings.follow up questions: (to formulate)

3.7.2 Full relative clauses (tensed finite relatives)

Do full relative clauses precede or follow the N, are both possible? or does theN internal to the relative clause?(follow up with the sswl relative clause properties).

(40) the fish that you should grill tonight„,

25

4 Combining adjectival fields: orders with re-spect to the N, relative orders of adjectives,and tables

4.1 Basic orders etc: Indefinites1. Choose adjectives from the different fields in your table below, (for the fieldsthat your language allows), and check if you can combine them with N;2. Vary the order of multiple As with respect to the N; check indefinites/definitesseparately),. Make sure where there is no contrastive interpretation, i.e. nocontrast).

subjective comment& size&age&shape&color& gender&nationality&materialbeautiful & small & new/old & round &red & female & French & wooden &&

Here is an example for some combinations of adjectives for indefinites nounphrases,

(41) Context: What are you thinking about?

a. A beautiful small tree (indefinite)b. An old French cityc. A big red birdd. A beautiful old French city

And for some combinations in definite noun phrases:

(42) Context. There is a beautiful small tree and a nice red bird on thispainting. Which one do you like best?a. The beautiful small tree

b. etc.

Indefinites: basic questions.

(43) a. Can you combine two As with the N? (a beautiful red ball)Yes/No/No sure(Give some examples)

b. Can you combine three As with the N?( This is not possible In some languages ) (yes/no, give examples)

c. If you can combine all the 7 regions, please do, and give an example(Even if these may be difficult to interpret, they sound fine.)Is the order in (neutral contexts) rigid, or variable? Is one orderprefered?

26

If the order is variable does it depend on the form of the adjective(predicative, adnominal)?If the order is variable, is this true for both indefinite and definitenoun phrases?

yes/no for indefinitesyes/no for definites

4.1.1 Table: General Order w.r.t to N in Indefinite noun phrases

& & yes & no &perhaps&N A A & & &&A A N & & &&A N A & & &

4.1.2 Table: General Order w.r.t. N: definites& & yes & no &perhaps&N A A & & &&A A N & & &&A N A & & &

4.2 Relative Ordering of adjectival regions: a full paradigmwith follow-up tables)

This section contains an example of a full paradigm of combinations of twoadjectival regions, and their relative ordering possibilities.For datable purposes it is important to break the data down into as many smallparts as necessary: to this effect, each section contains a summary table forthat combination. Simply check the relevant boxes.

Here is the task for your language. The examples below are based onEnglish orders

(44) a. If both A precede the N (A A N ) or follow the N (N A A), sys-tematically vary the orders of As (no contrast)

b. If A N A is yes: can the adjectives be reversed? (cf. example: abeautiful tree small/ a small tree beautiful) (no contrast)

What you may expect to find. All this will be coded in the tables after eachcombination.

• you may find that the order is rigid

27

• that there is a preferred order

• that both orders seem equally fine

• that you are unsure.

• that there is some difference which may be hard to in down.

• That this may vary for different combinations of adjectival fields (if thisis the case, see if there is a generalization that emerges, does it dependon the adjectival field and its position in the hierarchy? The form of theadjective? Are tue adjectives closer to the N than reduced relatives? etc.The environment in which the adjective occurs (bare versus linker?) etc...

4.3 Subjective comment with all other regions4.3.1 Subjective comment and size

(45) a. A beautiful small treeb. A small beautiful tree

& & yes & no & not sure1&Are both orders possible & & &*2&If 1 is yes, do the forms of the A vary & & &3&Is one order preferred & & &4& if 3 yes, which order?5&Is one order excluded? & & && if 5 yes, which order?

If 1 is yes, answer the questions below. Otherwise go to the next section.Follow up questions if row 2 is yes or no, or not sure.

• If 2 is yes:Is there a difference in form between adnominal and predicative adjectives?Yes/NoGive examples.

• If 2 is no, and if there is a difference between adnominal adjectives andpredicative onesare both forms adnominal A? Yes/Noare both forms predicative A ? Yes/NoIs one form adnominal and the other predicative (give examples)cannot tell?

• Sometimes, one order feels like a coordinated structure. If this is yourintuition, (perhaps because of the prosody), is this true for both orders,or for one order in particular? Which order (give example)?

28

4.3.2 Subjective comment and age

(46) a. A beautiful old tableb. An old beautiful table


If 1 is yes, answer the questions below. Otherwise go to the next section.Follow up questions if row 2 is yes or no, or not sure.




4.3.3 Subjective comment and shape

(47) a. A nice round tableb. A round nice table


If 1 is yes, answer the questions below. Otherwise go to the next section.

Follow up questions if row 2 is yes or no, or not sure.

29




4.3.4 Subjective comment and color

(48) a. A nice red flowerb. A red nice flower







30

4.3.5 Subjective comment and gender

(49) a. A beautiful male birdb. A male beautiful bird







4.3.6 Subjective comment and nationality

(50) a. A beautiful Malagasy womanb. A Malagasy beautiful woman




31




4.3.7 Subjective comment and material

(51) a. A beautiful wooden chairb. A wooden beautiful chair







32

4.4 Combining Size with the remaining regions4.4.1 size and age

(52) a. a big old houseb. an old big house







4.4.2 size and shape

(53) a. a big round bellyb. a round big belly



33





4.4.3 size and color

If they can be combined:

(54) a. a small red birdb. a red small bird






34


4.4.4 size and gender

(55) a. a small female birsb. a female small bird







4.4.5 size and nationality

(56) a. a small chinese vaseb. a chinese small vase


35






4.4.6 size and material

(57) a. a big iron chestb. an iron big chest






36


4.5 Combining age with remaining regions4.5.1 age and color

(58) a. an old red teashirtb. a red old teashirt







4.5.2 age and gender

(59) a. a young female birdb. a female young bird

37







4.5.3 age and nationality

(60) a.b.





38



4.5.4 age and material

(61) a.b.







4.6 Combining color with remaining regions4.6.1 color and gender

(62) a.

39

b.







4.6.2 color and nationality

(63) a.b.




40




4.6.3 color and material

(64) a.b.







41

4.7 Combining nationality with remaining regions(65) a.

b.







4.7.1 nationality and material

(66) a.b.




42




4.8 Definites noun phrasesYou want to repeat the same exercise for definite DPs, for each pair, and discussthe behavior in definite DPs .

(67) Context: There is a big red flower in a small white bottle.

a. I like the small white bottle a lot. (definite)




• If 2 is yes:Are the A marked with a definite article?


43


5 Order: Num A N5.1 Order: Num A N indefinite(68) Check where As can appear w.r.t. the Numeral (indefinites).

Different numerals (1, 2, 3 vs, 8, 12) may show different orders.add comments if this is the case.

& order & Yes & No & Example1 & Num A N & & &2 & Num N A & & &3.& N A Num & & &4.& N Num A & & &5*. & A Num N & & &6. & A N Num & & &

Follow up questions:

• if 5 is yes:

– For single As: (check different As (forms) from different regions): canall classes of A appear in the A Num N order?

– Can nonintersective adjectives precede the Num N ? (show with ex-amples)

– Can reduced relative clauses precede Num N (i.e. is the RedRel NumA order fine?) Yes/ No/not sure (give examples for yes and no)

– If 5 is yes: what are the ordering possibilities for 2 adjectives?Take adjectives from two different adjectival regions, where A1 comesfrom a region to the left of A2 in table 4.1. (For example: (beautifulA1, size A2), or any of the headings in section 4.2 .

(69) a. Is the order A1 A2 Num N possible? (give examples)b. Is the order A1 Num A2 N possible? (give examples)c. Is the order A2 A1 Num N possible? (give examples)d. Is the order A2 Num A1 possible? (give examples)

5.2 three adjectives(70) a. If you can get 3 adjectives before the N, give the order in which

they occur, and state if this order is variable or fixed.b. Same for post nominal adjectives?

44

5.3 Order: Num A N definiteWhat are the ordering possibilities in definite contexts?

(71) Context: there are four long ropes lying on the floora. Can you hand me the four long ropes ?

Fill in the table for definite Num A N orders: and add comments where

appropriate:

& order & Yes & No & Example1 & Num A N & & &2 & Num N A & & &3.& N A Num & & &4.& N Num A & & &5*. & A Num N & & &6. & A N Num & & &

Follow up questions:

• if 5 is yes: If row 1 and 2 are yes (both Num A N and Num N A arepossible in indefinite contexts

6 Orders: Dem A NThese correspond to already existing properties in SSWL.Check where As can appear w.r.t. Demonstratives. Give (or refer to) examplesfor all yes cases.

& order & Yes & No & Example1. & Dem A N & & &2. & Dem N A & & &3.& N A Dem & & &4.& N Dem A & & &5. & A Dem N & & &6. & A N Dem & & &

• If 5* is yes:

– form of the A? Can the A be a nonintersective adjective?– Can reduced relative clauses precede the demonstrative (RedRel Dem

A?– How do multiple adjectives behave?

Take adjectives from two different adjectival regions, where A1 comesfrom a region to the left of A2 in table 4.1. (For example: (beautifulA1, size A2), or any of the headings in section 4.2 .

(72) a. Is the order A1 A2 Dem N possible? (give examples)b. Is the order A1 Dem A2 N possible? (give examples)c. Is the order A2 A1 Dem N possible? (give examples)

45

d. Is the order A2 Dem A1 possible? (give examples)

Examples for yes cases:Follow up questions:

7 Universal 20: Greenberg (1965) and Cinque(2005): the relative orders of Dem Num A N

In this section, you are asked to look at the ordering of 4 regions: demonstra-tives, numerals, adjectives and nouns. Translate the following sentences intoyour language ( no contrast !), give examples of all possible orderings (in aneutral context).

(73) a. (I picked) these three beautiful flowersb. (Give me) those two red balls (there are two red balls) etc.

The 24 possible orders are split into several tables, depending on what elementscomes first. Fill in the tables, so that all variations and orders are recorded

7.1 Table Dem initialDem initial

&order & yes & no &not sure1&Dem Num A N & & &2&Dem Num N A && &3&Dem N Num A & & &4& Dem N A Num & & &5*& Dem A N Num & & &6*& Dem A Num N & & &

7.1.1 follow up questions Dem initial

• If 5 is yes: structural ambiguity: [A N ] two (’i.e. two [large elephants] )or A as reduced relative. A [ N Num ]Check if the A can be non-intersective, or if A must have a predicativeform. Can reduced relatives appear in this position in the language? (seeappendix)

• If 6 is yes, Check if the A can be non-intersective, or if A must have apredicative form. Can reduced relatives appear in this position in thelanguage? (see appendix)

7.2 Table N initialN initial

46

&order & yes & no &not sure1& N A Num Dem & & &2& N Num A Dem & & &3&N Dem Num A & & &4*& N Dem A Num & & &5*&N Num Dem A & & &6*& N A Dem Num& & &

7.2.1 follow up questions N initial

• If 4 is yes: check if A can be non-intersective, or if A must have a predica-tive form. Can reduced relatives appear in this position in the language?(see appendix). Give examples

• If 5 is yes: same questions as 4.

• If 6 is yes, same questions as 4

7.3 Table A initialA initial

&order & yes & no &not sure1& A N Dem Num & & &2& A N Num Dem & & &3*& A Dem N Num & & &4*& A Dem Num N & & &5*& A Num Dem N & & &6 *& A Num N Dem & & &

7.3.1 Follow up questions A initial

• If 3, 4, 5, 6 are yes: check if A can be non-intersective, or if A must havea predicative form. Can reduced relatives appear in this position in thelanguage? (see appendix). Give examples

7.4 Table: Num initial

Num initial&order & yes & no &not sure1*&Num A N Dem & & &2*& Num N A Dem & & &3*& Num Dem A N & & &4*& Num A Dem N & & &5*& Num Dem N A & & &6*& Num N Dem A & & &

47

7.4.1 Follow up questions Num initial

• If there are any Num initial orders:What do partitive constructions in your language look like? Are theyformally different from the Num Dem A N orders?Here is an example of a partitive construction

(74) Context: there are five nice children playing with a soccer balla. Three of these nice children are laughing, andb. Two of these nice children are singing

• Translate (74) this into your language, (keeping the noun phrases long).Is this order fine with different demonstratives? Can this order occur inclearly indefinite contexts in your language?

• If your language has an element like other, where would it appear in thepossible orders in your language?

7.4.2 Other

MOVE TO APPENDIXExpand:Order: other A N (6 possibilities). (+ Indef article or not)Order: other A N (6) (+ define ’article’ , + split indef article or not, + . Insome languages (Lebanese arabic), cannot occur in definite contexts.Order other and Numerals (indef/ definite and meanings)

two other boys / *other two boys the other two boys the two other boysForm: sometime a special demonstrative; sometimes an ordinal like second(Hindi)

8 To do: here is an incomplete list of stillmissing from questionnaire, and appendix:feel free to add using the formats given

• focus on A ( a BIG elephant, not a SMALL elephant)a big MOUSE , not a big ELEPHANT

• focus on numerals. (THREE big elephants, not TWO big ones)

• focus on Dem, and pronominal possessors. (widely attested in GrassfieldBantu )

• order: possessives, genitives (pronominal vs names vs DPs).

• ambiguities or not? Depends on order? In English, my former mansion isambiguous, between two readings:

48

& order & Yes & No & not sure1 & RedRel N & & &2 & N RedRel & & &

& order & Yes & No & not sure1 & modRedRel N & & &2 & N modRedRel & & &

(75) a. my former mansion (sa mansion formerly owned by me in thepast)

b. [ a former mansion] that I own. (no longer a mansion.

• ellipisis !

• expressions with light nouns: something big.

• etc

• Ordinals ! The first elephant .. This is a first start..

A MOVE: Further questions about reducedrelative clauses

A.1 Order: N red Rel/Red rel NIf reduced relatives can contain modifiers, the ordering possibilities may bedifferent. Check unmodified reduced relatives, and modified reduced relativesseparately

(76) a. I saw a burned house (unmodified) vs.b. I saw a [ completely burned out ] house (modified)

Do unmodified reduced relatives precede or follow the N? Or are both orderspossible?

NB: If more than one order is possible. there may be meaning differences,sometimes hard to pin down ! (See Cinque (2010, section 3.3. for furtherdiscussion and references.)

A.2 Order Num RedRel NWhat is the order of Numerals and reduced relatives in definite noun phrases?

(77) (On his way home,) he saw three burned houses

Give example.Fill in the table below. Check as many yes as apply.

49

Order: (RedRel = reduced relative)& order & Yes & No & not sure1 & Num RedRel N & & &2 & Num N RedRel & & &3. & N RedRel Num & & &4.& N Num RedRel & & &5.*& RedRel Num N & & &6. &RedRel N Num & & &

A.3 Order Num RedRel N IndefWhat is the order of Numerals and reduced relatives?

(78) (On his way home,) he saw three burned houses

Give example.Fill in the table below. Check as many yes as apply.

Order: (RedRel = reduced relative)& order & Yes & No & not sure1 & Num RedRel N & & &2 & Num N RedRel & & &3. & N RedRel Num & & &4.& N Num RedRel & & &5.*& RedRel Num N & & &6. &RedRel N Num & & &

A.4 Order: Ord RedRel N definiteFollow up on row 5: If 5 is possible, check the order also with ordinals anddemonstratives.

(79) a. On his way home, he saw his third burned-out house

Order: (RedRel = reduced relative)& order & Yes & No & not sure1 & Ord RedRel N & & &2 & Ord N RedRel & & &3. & N RedRel Ord & & &4.& N Ord RedRel & & &5.*& RedRel Ord N & & &6. &RedRel N Ord & & &

50

Table 7:

Order & yes & no & don’t knowAidiom N & & &N Aidiom & & &

A.5 Order Dem RedRel NOrder: (RedRel = reduced relative)

& order & Yes & No & not sure1 & Dem RedRel N & & &2 & Dem N RedRel & & &3. & N RedRel Dem & & &4.& N Dem RedRel & & &5.*& RedRel DemN & & &6. &RedRel N Dem & & &

B Compounds [SKIP? ELABORATE]Compounds

(80) Does your language have A N/ N A compounds?2 Yes/ No/ Don’t know

a. If yes, how are they ordered?

b. If you pluralize the N where does plural show up? on N, on bothN and A?

(81) Does your language have N N compounds. (Yes/ No/Not sure) book-shelfrooftopsick people house ’hospital’

(82) Does you language have ("er") V N or N V(er) compounds, or both

’church’– N V: house (of) pray (in); V N ’pray house’tango dancer or dancer tangodrum player or NCLass play drum ("er" play drum), or player (LNK) drum

2Follow up with Tamanji (april 14) on Bafut, both compounds (N A) (plural outside) andidioms (Plural on both

51

Table 8: Headedness compound

Order & yes & no & don’t know’er’-V N & & &V-er N & & &N V- ’er�

guided questionnaire for the development of a sswl type … · 2019. 5. 2. · 1 guided...

Documents